Grouping of a set of lines is operationally defined as common reversal of the lines under conditions of depth ambiguity. The degree to which this occurs can be measured, and constitutes a quantitative measure of perceptual grouping. Two grouping procedures, organization and coupling, are distinguished, and differences between them will be investigated. The effects of line separation, line orientation, hemispheric projection, context, line colinearity, and type of line intersection as determinants of grouping will be investigated. It is also proposed to evaluate the role of eye movements in grouping, the degree to which it is modifiable by set and instructions, and the level of processing at which it occurs. Experiments on the effect of head rotation on grouping and grouping on binocular rivalry are especially relevant to the latter problem.